Question regarding semantics

[Dragonfly]

You may assume that, but that doesn't take away the fact that you can define a new class by seeing only one object belonging to that class.

A class consisting of one object? I propose that this is meaningless. There is an implicit assumption of plurality in a class. We only know about one universe and we can speculate that their are other similar universes and go on to define what "a universe" means but it has no meaning if the universe is a singular event.

Now you're switching again to set theory, you're as slippery as an Objectivist...

The fact is that you wrote earlier:

The point is you cannot legitimately define a class of objects after having seen only one.

My example shows that you can do exactly that!

[tjohnson]

You may assume that, but that doesn't take away the fact that you can define a new class by seeing only one object belonging to that class.

A class consisting of one object? I propose that this is meaningless. There is an implicit assumption of plurality in a class. We only know about one universe and we can speculate that their are other similar universes and go on to define what "a universe" means but it has no meaning if the universe is a singular event.

Now you're switching again to set theory, you're as slippery as an Objectivist...

The fact is that you wrote earlier:

The point is you cannot legitimately define a class of objects after having seen only one.

My example shows that you can do exactly that!

Your example is not legitimate, it is meaningless. Also, I don't think 'slippery' is part of the definition of 'an objectivist'.

[BaalChatzaf]

GS, unfortunately our esoteric conversation is leading people in all sorts of weird directions.

Clarification on semantics: you cannot by definition have a class that contains only one object, correct?

How about the set of even prime integers?

Ba'al Chatzaf

What about it? The set of even primes is not an even prime.

The set of even primes has exactly one element, which is to say the CLASS or property; even primes applies to only one thing.

Have you every considered studying mathematics?

Ba'al Chatzaf

[tjohnson]

Have you every considered studying mathematics?

Ba'al Chatzaf

I have an honours degree in mathematics. Have you ever considered minding your own business?

[Xray]

Of course if you sense an object (say a tree) for the first time it goes without saying that you recognize a distinct object - my point is that you wouldn't call it 'a tree' until you observe multiple examples and invented the class name TREE. I think we agree on that.

I don't... If you discover a new animal, a single exemplar can be sufficient to give it a new name. Originally we knew only one sun and one moon, that didn't stop us to call then 'sun' or 'moon' (or equivalent names in whatever language you use). It was only much later that we realized that there are many similar things in the universe, although the name 'sun' is usually still reserved for 'our' sun (and 'star' for the other ones). There are many planets, but there is only one 'Earth'. So that whole theory of concept forming depending on perceiving multiple entities is nonsense.

Correct. For if I observe a distinct object, what's to keep me from attaching an audio/visual symmbol to this distinct object? I don't have to observe more than one of anything to label a singular.

Good examples DF with terms like "earth" and "moon" to expose Rand's theory of concept formation as the nonsense it is.

[Brant Gaede]

Have you every considered studying mathematics?

Ba'al Chatzaf

I have an honours degree in mathematics. Have you ever considered minding your own business?

The nastiest Internet flame wars are between mathematicians, I suspect.

--Brant

[tjohnson]

Correct. For if I observe a distinct object, what's to keep me from attaching an audio/visual symmbol to this distinct object? I don't have to observe more than one of anything to label a singular.

Good examples DF with terms like "earth" and "moon" to expose Rand's theory of concept formation as the nonsense it is.

If you see something you don't recognize then you may call it a name if you wish, like Bob, but this is not what I am talking about.

[merjet]

Good examples DF with terms like "earth" and "moon" to expose Rand's theory of concept formation as the nonsense it is.

Hogwash. As DF used it "Earth" is a proper name. There are multiple moons, not just the Earth's moon.

[Dragonfly]

Hogwash. As DF used it "Earth" is a proper name. There are multiple moons, not just the Earth's moon.

Perhaps in English the Earth is a proper name, but in Dutch for example it is a simple noun and not a name. The concept of a moon was known long before any other moons than our own were observed. The same for the concept "atmosphere" or some newly discovered species. There isn't any reason that you have to see more than one example of an object before you can form a concept of it. In fact there are many concepts of things that nobody has seen yet.

[tjohnson]

In fact there are many concepts of things that nobody has seen yet.

Notice that you used the word "things" in this sentence. Concepts or classes apply to groups. I accept that a class or concept can be created without seeing even one of the objects, however, it is assumed there will be more than one. If you find an example of a new species is it possible to have a species with one member?

[merjet]

Perhaps in English the Earth is a proper name, but in Dutch for example it is a simple noun and not a name. The concept of a moon was known long before any other moons than our own were observed. The same for the concept "atmosphere" or some newly discovered species. There isn't any reason that you have to see more than one example of an object before you can form a concept of it. In fact there are many concepts of things that nobody has seen yet.

Also in English "earth" (with lower case "e") is a general noun. For example, see this dictionary entry:

earth n.

5. the soft, granular or crumbly part of land; soil; ground

(source)

Arguing about whether or not two or more referents are needed for a concept is semantics without getting to the substance. Although Ayn Rand mostly used "concepts" instead of "universals" in Introduction to Objectivist Epistemology, she was quite clear that her main purpose was to address the "problem of universals" (ITOE2 1, 53, 74). In my opinion John Locke used the best descriptive label for universals -- general terms. On the other hand, if one uses "concept" as a synonym for "idea", then obviously one referent (which might be imaginary) is enough to have a concept.

Edited January 13, 2010 by Merlin Jetton

[BaalChatzaf]

earth n.

5. the soft, granular or crumbly part of land; soil; ground

(source)

Imagine that. We live on a world named Dirt.

Ba'al Chatzaf

[BaalChatzaf]

Have you every considered studying mathematics?

Ba'al Chatzaf

I have an honours degree in mathematics. Have you ever considered minding your own business?

I am surprised. No I haven't considered it.

Ba'al Chatzaf

[Christopher]

This conversation got sidetracked when Baal introduced sets - this is not about mathematics. The point I am trying to make is that we humans form classes or generalizations denoted by words like 'tree', for example, based on observations of multiple individuals. Also, we have a tendency, after producing these generalizations, to forget about the individual differences of the members of these classes. This can be very problematic for humans and it's a problem unique to us because of our advanced use of language compared to other animals.

GS, I think you're right in the realm of conceptual identification (which is the main subject, not math), and Merlin said it well - we are talking about universals; and of course a single exemplar can be the origin for defining a class To define a class off a single instance is to implicitly assume that more instances of the observed object exist. Additionally if I recall ITOE correctly, following Rand's logic if we observe three objects, two of which are blue and one which is red, we can only form an implicit concept for blue and not red. (implicit meaning automatic and without the use of imagination required to assume more instances)

Or as they learned in class one day at Oxford, a class of classes is not a class.... or is it?!!!!

Many brains explode on this fateful day of learning.

[Steve Gagne]

Chris, GS --

When Bob used the term "sets", I saw his point pretty clearly. (And I'm happy to be living on the Class-M planet Dirt.)

In earlier discussions I've pointed to "sets" of perceptions and experiences, probability, repeatability, and testability, as being the basis of learning and concept formation. A "set" is any envelope which one uses to integrate his percepts, as is "class". Beyond a slight difference in connotation ("set" would be descriptive; "class" would be normative), I do not understand the point of the discussion. Validity is not even part of either definition.

I see this discussion of "set" vs. "class" as being a distinction without a difference. So is this just a dispute about words? If not, please clarify.

Grax.

steve

Edited January 14, 2010 by Steve Gagne

[Xray]

If you see something you don't recognize then you may call it a name if you wish

That is, an arbitrary chain of sounds is attached to differentiate the perceived object from its surroundings. (e. g. "TREE"). In case the chain of sounds is accepted by a larger group, it can then be extended to label similar objects as well (category).

GS:We form concepts and definitions based on observations of multiple objects. It is not possible to form a class, like tree, unless we have observed many individual examples from which we abstract the concept tree from.....

What's to stop anyone from defining a single object? In fact, if a single object is not defined, how do you get to two or more single objects from which to select similarities to create a category?

Isn't a description of an entity a concept of the entity? What do you call it? Why object to calling it a concept? Isn't concept 'conceived idea of'?

Why do you assume that "definition" applies only to a word definition of a category?

Edited January 14, 2010 by Xray

[Xray]

On the other hand, if one uses "concept" as a synonym for "idea", then obviously one referent (which might be imaginary) is enough to have a concept.

Correct.

A concept is a "conceived idea of", which includes all ideas such as the idea of 'entity identity' which precedes the idea of category.

A category is a grouping by arbitrarily selected similarities.

In Rand think, it's all about only categories as concepts, as if there are no other ideas, no other concepts.

There may be categories of numerous entities, but the term, entity, per se means ONE.

In Randland, AN (one) entity identity (by difference) is not recognized.

Hence, the nonsense that concept refers to category only.

Edited January 14, 2010 by Xray

[tjohnson]

OK, you can have a concept of a particular tree but as Christopher pointed out, you need more than one to establish the validity of the concept of TREE, in general, with the idea that the any other objects of this group will also have the defining attributes.

[Christopher]

Chris, GS --

When Bob used the term "sets", I saw his point pretty clearly. (And I'm happy to be living on the Class-M planet Dirt.)

In earlier discussions I've pointed to "sets" of perceptions and experiences, probability, repeatability, and testability, as being the basis of learning and concept formation. A "set" is any envelope which one uses to integrate his percepts, as is "class". Beyond a slight difference in connotation ("set" would be descriptive; "class" would be normative), I do not understand the point of the discussion. Validity is not even part of either definition.

I see this discussion of "set" vs. "class" as being a distinction without a difference. So is this just a dispute about words? If not, please clarify.

Grax.

steve

A set (as used on this thread) is a mathematical concept. Conversely, when we talk about classes (on this thread) we are talking about how the mind derives conceptual categorization and labels aspects of that categorization process.

Hope this is the response you're looking for.

[Steve Gagne]

A set (as used on this thread) is a mathematical concept. Conversely, when we talk about classes (on this thread) we are talking about how the mind derives conceptual categorization and labels aspects of that categorization process.

Hope this is the response you're looking for.

Okay, I see why this is confusing to me. I don't use the term "set" as a strictly mathematical concept (and, and despite his background, I'm not sure Bob Kolker does either), but as a term to mean "any mental grouping of instances of pre-existing percepts or concepts". It is thus the basic unit of rational thought, and I suppose it is near-identical to your usage of the term "class".

But I can see why, if you regard it as solely a term limited to a specific field, you would regard its usage as irrelevant here.

As I stated earlier, to me the only distinction is connotative, and actually speaks to your concern about the categotization process. "Sets", the term with the merely descriptive connotation, can carry some sense of a seeming "randomness" concerning the collecting of percepts and concepts together, i.e., the integration process has not yet gotten to the stage of objective testing of integratives. Thus neither null sets nor single-instance sets can be excluded at this point. "Classes", on the other hand, seem to carry the normative connotation that some sort of objective testing and validation must have occurred before the conceptual unit is considered. So, does that require multiple instances? I don't know. We ARE talking about slightly different stages in the process, but still, the same multi-iterative process: perceive, isolate, identify, integrate, test, validate, iterate, act, a.k.a. concept formation, a.k.a. learning, a.k.a. induction.

So at what point do you see a "class" being formed, and why?

[Xray]

OK, you can have a concept of a particular tree but as Christopher pointed out, you need more than one to establish the validity of the concept of TREE, in general, with the idea that the any other objects of this group will also have the defining attributes.

What Rand means by concept is actually a lexical "category"; but how can one establish a category (of more than one) before first having a concept of an entity's identity?

Entity identity precedes categorizing. I doubt our stone-age ancestors waited before linking a sound chain to an object until they saw two or more.

Imagine a dialogue between them when they encoutered an unknown object for the first time: "No, we can't give this thing a name yet, we'll have to wait until we see at least one more!" In some cases, they could have waited forever ...

If Rand's theory were true, the sun and moon would not have been given any names until it was discovered centuries later that there exist more suns and moons.

My parents told me that as a small child, I attached arbitrary sound chains to some objects before I was even introduced to the sound chain conventionally used in my language to label the thing.

The German word for nutcracker is "Nussknacker". But I called this entity "hainaxan" at a very early age. This was not due to any phonetic misunderstanding of Nussknacker - I wasn't even aware the term 'Nussknacker' existed, nor was I aware that more than one nutcrackers existed in the world.

I used the arbitrary sound chain "hainaxan" in my relationship with this specific object, with this entity, to linguistically differentiate the object.

The funny thing is, the term 'hainaxan' which I created stayed in my family. My parents thought it sounded funny and even began using it jokingly themselves on occasions.

Lexicon entries like 'tree' refer to a linguistic category. Defining 'tree' is defining a term labeling a category.

What Rand did not take into account: categorizing is a completely arbitrary process. For there is no 'objective' reason to categorize the objects which are called "tree" in English in a separate class at all.

A language could for example have only one term, one linguistic category referring to all objects growing out of the ground: let's invent the term "kalawamba" for this. "Kalawamba" may then be (again, arbitrarily) subcategrized in e. g. those objects containing wood, which would comprise trees and shrubs. Or in eatables an non-eatables. The possibilities of categorizing are limitless.

The objects themselves exist independently of all that arbitrary categorizing and classifying.

Rand erroneously believed that getting an idea of an arbitrary linguistic category/class (she calls this "concept formation") was a complicated mental feat and connnected to any deep philosophical insight.

Describing a category is never defining the real object.

In our semantics class (many moons ago, when structuralism still 'reigned'), I recall us having to place so-called semantic markers with (+ or -) to describe linguistic categories.

"Woman" was for exampe: +human, +female , +adult; "girl" was +human +female -adult.

Functioned with some concretes, but you could forget it with terms like. e.g. "gentleness" and countless others. The limits of structuralism were quickly reached there.

Rand stayed caught in her categorizing bubble all her life.

It led her to believe for example that "love" could be obejctively measured (!); she tried to establish a hierarchy of terms like "liking", then the next stronger was "affection" and then the whole thing culminates in "romantic love".

She delusionally attributed to these terms an objective quantity like degrees have in a temperature scale.

Again, she confused arbitrary linguistic categories with the real.

Flaubert wrote a great novel about a similar error: Madame Bovary, who, when reading chivalrous romance novels in her youth, became so obsessed with a specific illusion of romantic love that she tried to carry it over into her daily life. It ended in a tragedy.

Rand's "man" in "life proper to man", again it refers to class, to category only. The real finite individual is abandoned and the category, the mental construct presented as the real instead. All totalitarian ideologies operate by the same principle.

Edited January 18, 2010 by Xray

[merjet]

If Rand's theory were true, the sun and moon would not have been given any names until it was discovered centuries later that there exist more suns and moons.

Hogwash. Do you enjoy repeating your errors after they have been recognized, like here?

Rand's "man" in "life proper to man", again it refers to class, to category only. The real finite individual is abandoned and the category, the mental construct presented as the real instead. All totalitarian ideologies operate by the same principle.

Do you see any difference between (1) somebody saying "you must ..." and (2) somebody saying "you must ..." while pointing a gun at you?

[Xray]

If Rand's theory were true, the sun and moon would not have been given any names until it was discovered centuries later that there exist more suns and moons.

Hogwash. Do you enjoy repeating your errors after they have been recognized, like here.

You still have not grasped what it is about. Per Rand, to form a "concept", it takes two or more objects from which similarities are abstracted.

But if a single object is not defined, how can you get to two or more single objects from which to select similarities to create a category?

MJ: Although Ayn Rand mostly used "concepts" instead of "universals" in Introduction to Objectivist Epistemology, she was quite clear that her main purpose was to address the "problem of universals" (ITOE2 1, 53, 74).

It looks like those so-called "universals" are (lexical) categories.

MJ: On the other hand, if one uses "concept" as a synonym for "idea", then obviously one referent (which might be imaginary) is enough to have a concept.

Correct. Rand creates confusion by using the term "concept" exclusively for linguistic category.

Xray: Rand's "man" in "life proper to man", again it refers to class, to category only. The real finite individual is abandoned and the category, the mental construct presented as the real instead. All totalitarian ideologies operate by the same principle.

MJ: Do you see any difference between (1) somebody saying "you must ..." and (2) somebody saying "you must ..." while pointing a gun at you?

Your point being? The one holding the gun is simply carrying out in practice what the ideologist has written in theory. Have you ever done the mental exercise to imagine what would have happened if Ayn Rand had been in a position of political (gun) power to have her ideas carried out?

Do you believe she would have accepted individualists telling her: "I want no part of it?"

Edited January 19, 2010 by Xray

[merjet]

You still have not grasped what it is about.

Ditto.

MJ: Do you see any difference between (1) somebody saying "you must ..." and (2) somebody saying "you must ..." while pointing a gun at you?

Your point being? The one holding the gun is simply carrying out in practice what the ideologist has written in theory.

Just what I said. You see no significant difference between Ayn Rand's ideology and a totalitarian one.

Edited January 19, 2010 by Merlin Jetton